Electrical stimulation over selected muscle tendons in alert human subjects produced, in each muscle, a reflex inhibition of muscle activity. This inhibition, when maximal, was seen in the surface EMG as an interval of complete electrical silence during a sustained voluntary contraction. The inhibition was clearly visible in single sweeps and in averaged records. Its onset latency and duration were respectively, 56 +/- 4.9 and 46 +/- 11.8 ms in extensor digitorum communis, 71 +/- 6.1 and 46 +/- 10.5 ms in extensor pollicis brevis, 77 +/- 11.2 and 47 +/- 10.5 ms in extensor pollicis longus, 72 +/- 7.3 and 43 +/- 8.6 ms in abductor digiti minimi, and 97 +/- 3.5 and 43 +/- 2.8 ms in tibialis anterior. The inhibitory response was produced at low stimulus intensities (< 10 mA) without electrical (M wave) or mechanical (muscle twitch) signs of direct muscle stimulation. It therefore did not arise from stimulation of la afferents (muscle spindles). The response arose from tendons since it occurred at lowest threshold when stimulation was applied directly over the tendons of the five different muscles studied. At low stimulus intensities, the response declined sharply when the stimulating electrodes were moved to the skin immediately adjacent to the tendons. The response did not arise from skin afferents since it was also presented when stimuli were delivered to the tendon by subcutaneous needle electrodes and it was not reproduced by stimulation of cutaneous nerves in the region of the tendon. In another series of experiments on extensor pollicis brevis, five skin locations were stimulated while overlying the tendon and again while the skin was stretched so that they were lying 0.6-0.8 cm dorsal to the tendon. In these experiments the response was again greatly attenuated when the stimulation was not directly over the tendon, although the same cutaneous sites were stimulated. The inhibition was followed by a pronounced excitatory component (E1) of peak latency 120-140 ms. The results of the study provide evidence for a powerful autogenic inhibitory reflex in man. The evidence is consistent with the possibility that the response arises from Golgi tendon organ afferents.